Several clinically important beta-lactam antibiotics are produce through condensation of D-(-)-alpha-aminophenylacetic acid with penam and cephem nuclei in multi-step chemical processes. An efficient enzymatic process for the synthesis of these antibiotics would lead to increased product yields and decreased cost of production. Previously described enzymes catalyze the synthesis of these antibiotics at rates that are inadequate for commercial processes. The aim of Phase I research is to demonstrate that enzymes able to catalyze the synthesis of these antibiotics at high rates can be derived from penicillin amidase. To achieve this, mutants will be selected following mutagenesis of cells that carry the cloned gene for penicillin amidase based on an improved ability to hydrolyze a model substrate with a D-(-)-alpha-aminophenylacetyl moiety. The rates at which amidase purified from these mutants catalyze the hydrolysis of the model substrate and the condensation of D-(- )-alpha-aminophenylacetic acid with 6-aminopenicillanic acid to produce ampicillin will be determined and compared to the rates observed with the wild-type amidase. The aim of Phase II research will be the reiterative use of the methods developed in Phase I to obtain enzymes able to synthesize ampicillin and cephalexin at commercially useful rates.